My purpose in buying a Pi-top is to have a convenient way to use the Raspberry Pi desktop. I am porting some educational software, and while for most purposes a Pi on the network is sufficient, accessed via ssh, sometimes I need to see what things look like on the Pi screen.
First impressions are good as regards design and ease of use, but not perfect where quality control is concerned. The keyboard, touchpad and screen are all comfortable and easy to use -- in fact, all three are better than I expected. I wouldn't want to type a novel on the keyboard instead of my beloved thinkpad, but it is good enough for me to type without noticing it all the time. The trackpad supports multitouch gestures and tap-to-click in a seamless way. The display is nice and big, and plenty bright enough.
Quality issues affected the insides of the case more than the outsides. Although the fit of the panels on the outside is not perfect, it is good enough, and I don't see that it could be much better given that it snaps together and the keyboard slides out. Inside, however, the story was not so good on the machine I received. The two magnetic rails are glued in place, and not enough glue had been used on one of them, so that it was loose. The fit of the Raspberry Pi on its mounting holes was far enough out that is was difficult to fit the mounting screws. And worst of all, the threaded brass insert in the case for the screw that holds down the cooling bridge was loose, so that it pulled out of the chassis as soon as I tried to tighten the screw. Fixing this entailed disassembling the entire sliding PCB arrangement, and probably led to the fourth problem: that the keyboard and touchpad didn't work on initial boot. I fixed this by reseating the ribbon cable for the keyboard which (disappointingly for a machine intended for inventors) has a 'Warranty void' seal on it. Luckily, the cooling bridge seems to stay in place quite well without the screw.
Overall, I am happy with the machine so far. I didn't spend long looking at the bundled software, and won't bother with the ho-hum collection of LEDs and cardboard cutouts that make up the 'inventor's kit'. As a bonus, it would have been nicer to have a speaker or a real time clock, or even a solderable breadboard in place of the tiny solderless one. I will no doubt rig up an RTC in some way soon. But stock Raspbian Stretch booted without issue, the screen brightness control worked out of the box, and it was easy to install a widget to monitor the battery.
Yes, there is a small high-pitched noise but when you close the pi-top the noise is lower.
I've heard a high-pitched noise too, so faint I'm sure you wouldn't hear it if there was any kind of cooling fan. I haven't tracked down where it's from, but I suspect the screen and the big inductor on the hub board. I'll report if I manage to track it down.
When the pi-top speakers version 2 are available, you will hear the sound of your speaker(s) only after installation of the speakers.
Thanks a lot for that, indeed it does stop when going full brightness.
Yes, I'm finding the noise (always faint in my case) goes away entirely when the screen is set to full brightness. If they could double the PWM frequency for the backlight, it would bother only passing bats: the ear can't hear as high as that!
Probably showing my age, but I don't hear a thing.
No noise my end. Makes me wonder what the issue is since there are no moving parts.
My guess is that the backlight brightness is adjusted using pulse-width modulation, an attractive method because it saves energy and minimises dissipation in the electronics. Some component, perhaps the screen itself, is acting like a loudspeaker, emitting sound waves at the PWM frequency and its hamonics. The PWM frequency itself (or maybe a sub-harmonic) is just barely audible (depending on your age) but the harmonics are not. When the brightness is near maximum or minimum, the noise seems quiet because the pulse width is very wide or very narrow, containing little of the fundamental frequency and lots of harmonics. At mid-range, the noise seems louder because more of the fundamental frequency is emitted.
All this is just a guess, of course, but it's consistent with what we are finding. As I remarked above, if the PWM frequency were doubled, probably nobody would be able to hear it.
An update on the mechanical problems;
I had half an hour to spare and a soldering iron to hand, so I decided to try fixing the problem with the loose brass insert in the case. Holding the iron on it for a few seconds, and squeezing the sides of the boss with a pair of pliers, seems to have fixed the problem quite easily. The screw for the cooling bridge now goes in easily and seems quite secure.
On reassembling the machine, however, the problem with non-working keyboard and trackpad recurred. This time I had stock Raspbian Stretch installed on the SD card, and messages about "not being able to read the descriptor" for certain USB devices appeared below the row of four raspberries before the boot splash screen appeared. I tried reseating the ribbon cable again, but to no good effect.
Then it struck me that the keyboard was more likely connected via one of the captive USB plugs that connects to the Pi than by the ribbon cable. I tried removing the RPi and flexing these upwards with the idea of making a firmer connection. I was able to establish with a multimeter that one of the plugs had four pins in continuity with the pins of the internal USB connector on the hub board, and that the other had power and ground connected, but couldn't find a way of testing for continuity of its data lines. Reassembling gave a machine that would sometimes work, and I subsequently swapped the RPi with another board that seems to have a better set of USB sockets. Everything is now working well, touch wood!
Lesson learned: if keyboard and trackpad are not working, suspect the connection of the captive USB plugs to the RPi. I can't prove that was the problem (that's the nature if intermittent faults), but it seems very likely. The ribbon cable was just a red herring, but playing with it just moved the USB connections enough to make them work for a while.
I have discovered when the keyboard and pad are not working, the change they work forever is to shut down the system. ( they must be worked before when you do this; connections inside where good in former time ) ; afterwards you must disconnect the internet UTP/FTP cable from the RJ45 connector.
When reconnected and the system is boot you have sometimes luck that the system will work good. 14 days further now and this laptop is working great. And i have sound to from the new speaker from modmypi.